[cle-release-team] [sakai2-tcc] Emerging consensus to focus on 2.10

[Noah Botimer]

OK, let me start with where we stand together -- this is a fantastic idea, and I think you are right on, except in that we should delay anything. I agree that the numbers don't matter much, but...


I'm sorry, but I feel that a BIGGER AND BETTER single release is precisely the wrong approach. Doubling down to put more stuff in a release is a major problem, resulting in less frequent releases, later and less QA coverage, and slower adoption.

We have suffered this for major and maintenance releases for years. They build inertia and get harder and harder to release and harder to adopt when we get there. It even delays and kills ambitious projects waiting for a time when they can be disruptive.

I do think a big-M release is a good idea, but it seems rather separate from the 2.10 schedule. There is already stuff in trunk that should be released. If there is other great stuff ready by a reasonable time to release the incrementals, that'd be wonderful. I'm just saying that being ambitious and releasing valuable, less ambitious stuff are not mutually exclusive.

I stand by it that I think a "same as expected" 2.10 release that tightens up 2.9 and a "whoa, that's awesome" 4.0 release is the winning strategy.

I think Apple has done this well since Leopard and it bears out that adopters are divided by their contentment with something familiar and solid versus picking up something that is different and has more features. If we did what I suggest, I think we would find a lot of 2.10 and 4.0 adopters, spread out pretty naturally.

I also stand by it that 4.0 should be bold and move to Github/Bitbucket and Gradle with a reduced core size (an easier build/install process making contrib a fine place to live) -- note that both can happen readily in parallel to preparing a 2.10 release.

I think we are very close together in direction -- just my same old drumbeat of "work on a pragmatic, imminent release and an ambitious, future release simultaneously".

Thanks,
-Noah



P.S., I'm on 10.6.8 now and I think it'd be fine if someone ran 2.10 for three years. When I get new hardware or there is something I can't resist, I'll jump to the front. I think institutions run the same calculation when there is a big feature or technical bridge to cross for an upgrade. It's natural and we should embrace it.

On Sep 6, 2013, at 7:09 AM, Steve Swinsburg wrote:

> Anything from the massive list of stuff we have collected, from the past two conferences TCC meetings etc.
> 
> On Fri, Sep 6, 2013 at 10:34 AM, Neal Caidin <neal.caidin at apereo.org> wrote:
> What specifically would you like to see in a Big M release?
> 
> On Sep 5, 2013, at 7:19 PM, Steve Swinsburg <steve.swinsburg at gmail.com> wrote:
> 
>> I think this could tie in nicely with the CLE 4 (or whatever version) discussion. Why don't we put a hold on getting 2.10 out the door, and take some additional time to invest in a Major (big M) release, targeted for say the end of 2014. 2.9 is in a good place, it can still have a few maint releases here and there during this timeout we can focus on really big items to get into the next Major release.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/cle-release-team/attachments/20130906/043bdadf/attachment.html